The Heritability of Abdominal Aortic Aneurysms Versus Thoracic Aortic Aneurysms

Approximately 20% of thoracic aortic aneurysms are associated with an autosomal dominant gene variation. This includes the genes for fibrillin-1 (Marfan syndrome) and the transforming growth factor (TGF)-β receptors TGFB1 and TGFB2 (Loeys–Dietz syndrome) as well as several genes involved in smooth muscle cell contractility. These genes have been identified by linkage analysis and gene sequencing in extended families with thoracic aortic aneurysms noted in several generations. There are few such families with AAA in many generations, so that identification of specific genes associated with AAA has proved much more difficult.

Certainly there is little evidence that genes involved in smooth muscle contractility are important in AAA. One of the reasons for this difference might be the different embryologic origin of the smooth muscle cells in the thoracic and abdominal aorta. The smooth muscle cells of the descending thoracic aorta are derived from somites, and the smooth muscle cells of the abdominal aorta are derived from splanchnic mesoderm. Although AAAs do occasionally occur in patients with thoracic aortic aneurysm genetic syndromes (e.g., Marfan syndrome), these are rare causes of AAA.

Inheritance of Abdominal Aortic Aneurysm

The observation that genetic factors have an etiologic role in the development of AAA has led researchers to study the nature and mode of inheritance of AAA. This information is helpful because it might demonstrate a mode of inheritance that can narrow down the likely genetic location of genetic variants responsible for AAA. This information can then be used to target detailed genetic investigation. The approach that has been adopted in most studies of AAA consists of the AAA family pedigree analysis.

Several authors have conducted pedigree analyses of AAA, but the results among studies are not consistent, with different authors suggesting either an autosomal monogenic or multigenic model of inheritance, recessive, or monogenic with gender-dependent penetrance. One potential reason for these inconsistencies is the limited size and generation span of the families investigated. Another potential reason is the rapidly increasing life expectancy of men in Western society, with wars no longer removing more men than women at ages before aneurysms become manifest; after all, AAA is principally a disease of older men.

The inconsistencies among individual studies make it difficult to draw robust conclusions. To improve upon these studies, Kuivaniemi and colleagues combined all of the data from these studies and additional data to generate a dataset of 233 AAA family pedigrees from seven countries (Figure 1). Multiple different modes of inheritance were again observed. A high percentage of autosomal recessive inheritance patterns were observed (72%) but, in addition, many pedigrees demonstrated autosomal dominant inheritance patterns, and in a small percentage of pedigrees autosomal dominant inheritance with incomplete penetrance was observed. These data led the authors to conclude that the genetic contribution to the development of AAA was multifactorial, and that there was a significant environmental effect in addition to the genetic factors.

Identifying Genes Associated with Abdominal Aortic Aneurysms

Given the aforementioned evidence, much research into the genetics of AAA has focused upon the identification of genetic variants associated with AAA. Evidence from other multifactorial genetic diseases (coronary heart disease, inflammatory bowel disease) suggests that each of the genetic factors contributing to the overall genetic variation causing disease have only small individual effects. These small effect sizes make the identification of the true variants difficult because experiments to identify such factors are prone to experimental error.

The principal experimental approach to determining the genetic variants associated with AAA has been to conduct association (case-control) studies. If the incidence of any particular genetic variant in patients with AAA is significantly different from the incidence of the same variant in healthy controls, then this suggests that this variant is associated with AAA. The choice of variants to be examined in such studies is often made based upon biologic plausibility for a role in AAA (such as polymorphisms of the matrix metalloproteinase [MMP] genes). This candidate gene approach is attractive because it enables the testing of a specific hypothesis. However, case-control studies, particularly small ones, are subject to confounder bias, and this has been ignored in many studies. In addition, there is the problem of publication bias: Many studies showing negative results are not accepted for publication.

Several candidate-gene association studies of AAA have been performed but usually in small, underpowered studies, and these findings are rarely confirmed in larger, adequately powered studies. For example, a promising association in the interleukin-10 gene promoter region identified in a case-control study of 100 patients with AAA and 100 controls was not associated with AAA when analyzed in a larger case-control study adjusted for socioenvironmental risk factors. A meta-analysis of candidate-gene association studies has provided convincing evidence for only three previously identified associations. This meta-analysis has determined that genetic polymorphisms in the angiotensin-converting enzyme (ACE), methyltetrahydrofolate reductase (MTHFR), and MMP9 genes are associated with AAA, with risk ratios of 1.33 (95% confidence interval [CI], 1.20–1.48), 1.14 (1.08–1.21) and 1.09 (1.01–1.18), respectively. One variant on chromosome 9p21, originally identified as a genetic association with coronary artery disease, also held an association with AAA in two large independent association studies. However, the extent to which this association is independent of coexisting coronary artery disease in patients with AAA is unknown.